EXPANDING THE LIMITS OF WOOD POLYMER COMPOSITES: STUDIES USING DYNAMIC MECHANICAL THERMAL ANALYSIS. W. Dale Ellis* and Anand R.

Transcription

1 Proceedings of the 18th Risø International Symposium on Materials Science: Polymeric Composites - Expanding the Limits Editors: S.I. Andersen, P. Brøndsted, H. Lilholt, Aa. Lystrup, J.T. Rheinländer, B.F. Sørensen and H. Toftegaard. Risø National Laboratory, Roskilde, Denmark, 1997 EXPANDING THE LIMITS OF WOOD POLYMER COMPOSITES: STUDIES USING DYNAMIC MECHANICAL THERMAL ANALYSIS W. Dale Ellis* and Anand R. Sanadi** *USDA-FS, Forest Products Laboratory 1 One Gifford Pinchot Dr. Madison, WI USA **College of Agriculture and Life Sciences University of Wisconsin Russell Laboratories Madison, WI USA **Visiting Associate Professor Department of Economics and Natural Resources Plantfiber Laboratory The Royal Veterinary and Agriculture University DK-2630 Taastrup, Denmark ABSTRACT Wood-Polymer Composites (WPC) are unique composite materials where a natural composite (wood) is impregnated with monomers that are then polymerized in the wood to tailor the material for special applications. The polymers are used to enhance specific properties of wood such as dimensional stability, hardness and abrasion resistance. The glass transition and secondary transition temperatures of the impregnated polymer are important parameters when considering the performance of polymer composites. In our experiments, using formulations of acrylic based polymers, the transition temperatures of WPC have been characterized using dynamic mechanical thermal analysis (DMTA). A combination of acylic monomers and other components were used to make polymers having differing properties. The effects of these polymer systems on transition temperatures measured by DMTA of the composites are presented. The effects of the interaction of the polymers with wood (the interphase) are discussed. Dynamic properties of the composites are related to physical properties of the composites. 1 The Forest Products Laboratory is maintained in cooperation with the University of Wisconsin. This article was written and prepared by U.S. Government employees on official time, and it is therefore in the public domain and not subject to copyright. 307

2 Ellis and Sanadi 1. INTRODUCTION Dynamic Mechanical Thermal Analysis (DMTA) is a useful method of detecting relaxations in polymer molecules as the temperature is scanned over a range from subambient to above the material s glass transition. This method of analysis is much more sensitive than differential scanning calorimetery (DSC) for the detection of molecular relaxations at the glass transition temperature (Tg) and can detect and measure secondary transitions (Gearing and Stone, 1984) which are nearly impossible to detect using DSC. Viscoelastic relaxations of chemically treated wood have been studied using a dynamic viscoelastomer (Sugiyama, Obataya and Nonmoto, 1995). The polymerization of glycidyl methacrylate in wood has been studied, the course of the polymerization was followed by measurement of the tan δ (Balaba and Subramanian, 1986). We have studied the temperature dependence of the loss tangent tan 6 of wood polymer composites (WPC). WPC are solid wood impregnated with monomers subsequently polymerized in the wood. Our objective was to characterize the changes of tan δ as affected by monomer combinations and by the interaction of monomers/polymers with the wood. Several factors can affect the interactions in the polymer impregnated wood. These include: 1. hydrogen bonding of the polymer with wood (hydroxyl groups) 2. covalent bonding to the wood 3. the presence of a surface can also result in different molecular weight distributions and crosslink densities in the interphase region of the interactions. In general, an increase in the transition temperature usually indicates the restricted mobility of the polymer molecules in the interphase region of the interface and is usually indicative of good interaction. Either one or a combination of any of the above factors can be contributing to this behavior. The use of DMTA as a technique for evaluating properties of WPC was explored by looking for relationships, if any, of tan δ with the hardness of the WPC. 2. EXPERIMENTAL 2.1 Monomer solution prearation The monomers used to prepare specimens set 1 were isobornyl methacrylate (IBMA), glycidyl methacrylate (GMA), hydroxyethyl acrylate (HEA) and hexanediol dimethacrylate (HDDMA). The monomers were used individually and mixed in molar ratios for the preparation of specimens (Table 1). The base triethylamine (TEA) was used to catalyze the opening of the oxirane ring in GMA. Specimens for set number 2 were prepared using the monomers isobomyl methacrylate (IBMA) and neopentyl glycol diacrylate (NPGDA) mixed on weight to weight ratios (Table 2). All the monomer solutions contained Vazo 52, a free radical initiator. 308

3 Studies using dynamic mechanical thermal analysis 2.2 Specimen Preparation. Maple veneers about 1.0mm thick and 100mrn by 100mm were impregnated with acrylate or methacrylate monomers subsequently polymerized in the wood. The untreated veneers were placed in a vacuum chamber, a vacuum applied, after 30 minutes of vacuum the monomer solution was drawn into the chamber to cover the wood. The chamber was then returned to atmospheric pressure. The veneers were soaked in the monomer solution for one hour then removed and the excess monomer wiped from the surfaces. The veneers were placed in a press heated to 100 C to polymerize the monomers. All specimens were post cured for 2 hours in an oven at 120 C. Excess polymer was removed from the surfaces of the veneers by light sanding. Specimens for DMTA were cut from the center section of the 100mm square veneers. They were cut 5mm wide by 20mm long and 1mm thick. These specimens were dried at 105 C then stored over a desiccant until analyzed. 2.3 Dynmanic Mechanical Thermal Analysis. Specimens were analyzed using a Rheometrics model Wk III dynamic mechanical thermal analyzer. A small single cantilever holder was used to analyze the specimens in a bending mode. A frequency of 1 Hz and a strain of 1% was used. The temperature was programmed at 2 degrees C per minute from -50 C to 300 C. 3. RESULTS Monomers were selected because of known properties of their polymers. 1. IBMA polymer has a high Tg relative to HEA polymer. 2. The hydroxyl groups of HEA polymer can potentially interact via hydrogen bonding with the constituents of wood. 3. The oxirane ring of GMA can make covalent bonds with groups in wood. 4. HEA polymer can be crosslinked with the dimethacrylate monomer HDDMA. IBMA can be crosslinked with the diacrylate NPGDA. The tan δ plot from DMTA was used to evaluate changes in transition temperatures (Tg) that could give some understanding of polymer-wood interactions. 3.1 Crosslinking within polymers. WPC specimens made using IBMA, GMA or NPGMA appear to have little or no molecular interactions with the wood to change the major transitions. The observed transitions are close to the known transitions of the bulk polymers. The difiunctional acrylate NPGDA used to crosslink KBMA gives polymers with higher Tg than that of IBhA polymer. IBM/NPGDA mixed 2:1 and 1:1 by weight raises the Tg from 113 C for IBMA without NPGDA to 120 C and 126 C respectively (Table 3). HEA polymer in wood has a Tg of 21 C, slightly higher than reported Tg of 15 C for bulk HEA polymer. This increase could be caused by the interaction of hydroxyl groups in the polymer with chemical constituents of the wood. We need to determine the Tg of bulk HEA polymer using conditions similar to our WPC analysis by DMTA before a more positive conclusion is made. An 309

4 Ellis and Sanadi increase of the Tg is seen when HEA polymer is crosslinked with 0.1 mole of the difunctional methacrylate HDDMA per mole HEA (Fig. 1). The resulting WPC specimen has a Tg of 38 C (Table 4). 3.2 GMA crosslinked polymer to wood. Relaxation in the polymers of WPC were observable as peaks in the tan δ versus temperature plot using DMTA. The tan δ Tg of 108 C for WPC prepared using IBMA was near the reported Tg of 111 C for IBMA. WPC prepared using IBMA/GMA and triethylamine to open and react the oxirane ring has the major transition 33 degrees higher than that of WPC specimens prepared using lbma/gma without triethylamine (Fig. 2). This increase of the transition temperature is seen as an indication GMA has attached (crosslinked) the polymer to the wood. WPC specimens prepared using GMA without triethylamine have a major transition at 89 C. WPC specimens prepared using GMA with triethylamine have no major transition peaks in the tan δ plot until 243 C (Fig. 3) which is the point the wood begins to degrade. It appears the GMA polymer has crosslinked with the wood sufficiently that the transition has broadened so much it is no longer 310

5 Studies using dynamic mechanical thermal analysis easily detected on the tan δ plot. We will do further experiments to confirm the attachment of polymer to the wood. 311

6 Ellis and Sanadi 3.3 Transition temperatures related to hardness of WPC. The Rockwell hardness(scale L) of WPC specimens is generally greater for the specimens having higher transition temperatures (range from 89 C to 142 C) than for the specimens having lower transition temperatures (range from210 to 96 C). The WPC specimens made with HEA in the monomer solution had lower hardness and lower transition temperatures than those of other WPC specimens (Table 4). The hardness of WPC is generally related to the hardness of the polymer but might also be affected by polymer-wood interactions. ACKNOWLEDGEMENTS One of the authors, A. R. Sanadi, would like to thank the Forest Products Laboratory for providing space and facilities for the study. Partial finding from the Forest Products Laboratory is appreciated. Authors would like to thank Alfred Christianson for helping using the DMTA. A. R Sanadi would like to thank Professor Per Olesen for arranging the stay in Denmark. REFERENCES 312